| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| NQ | DUC (Duplicatus Charm) | RepeatScore11 | 15 | 5mo ago | |
| AQ | RepeatScore0.53 | 15 | 5mo ago | ||
| EQ | RepeatScore0.46 | 15 | 5mo ago | ||
| BadCase 3 (test) | Direct Preference Optimization (DPO) fine-tuning | Repetition Rate Before45 | 1 | 5mo ago | |
| BadCase 2 (test) | Direct Preference Optimization (DPO) fine-tuning | Repetition Rate Before60 | 1 | 5mo ago | |
| BadCase 30 cases 1 (test) | Direct Preference Optimization (DPO) fine-tuning | Repetition Rate After0 | 1 | 5mo ago |
